نوع مقاله : مروری
نویسندگان
گروه شهرسازی،دانشکده هنر و معماری، دانشگاه یزد، یزد ، ایران
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Extended Abstract
Introduction
In recent decades, the rapid trends of urbanization and urban development have posed significant challenges for urban planning and management. These challenges include the management of resources and infrastructure, the quality of life of citizens, and the preservation of the environment. Given the accelerating growth of cities, identifying the patterns and key factors influencing urban development is essential for predicting and improving future trends. The application of machine learning algorithms—recognized as a branch of artificial intelligence—has created new opportunities in the analysis and modeling of urban development.
Urban development is influenced by various economic, social, and environmental factors, and accurately predicting it is of great importance to policymakers and urban planners. Despite technological advancements and the increasing availability of big data, analyzing and modeling these trends remain challenging. Integrating multidimensional data—including spatial, social, and economic data—and analyzing them through machine learning methods can provide a deeper understanding of urban development.
The primary objective of this study is to evaluate the effectiveness of various machine learning algorithms in analyzing urban development directions and identifying the associated challenges and opportunities. Specifically, this research aims to:
Assess the accuracy and performance of different algorithms in predicting urban development.
Analyze the integration of machine learning algorithms with Geographic Information Systems (GIS) and satellite imagery.
Investigate the challenges and limitations of using these methods in urban studies.
Methodology
In this study, MAXQDA software was used to conduct a systematic content analysis of research articles related to the analysis of urban development directions using machine learning algorithms. After preparing and importing the selected articles into the software, the coding process was carried out in three stages: open coding, axial coding, and selective coding. Using MAXQDA’s qualitative and quantitative analysis tools—such as Word Frequency, Code Matrix Browser, and Comparison Diagram—key concepts, relationships between codes, and methodological differences were examined.
To ensure the validity and reliability of the analysis, peer review and Cohen’s Kappa coefficient were employed. Finally, the findings were organized into themes and patterns, and compared with the existing literature to develop a conceptual framework for analyzing urban development.
Results and discussion
The results of this study indicate that algorithms such as Random Forest (RF), Artificial Neural Networks (ANN), and Support Vector Machines (SVM) play a significant role in predicting urban development. Integrating these algorithms with tools such as Geographic Information Systems (GIS) and satellite image analysis has enhanced modeling accuracy. However, challenges remain, including data quality issues, appropriate parameter selection, and the computational complexity of these models.
Conclusion
This study highlights that machine learning can serve as a powerful tool for forecasting and analyzing urban development trends. Nevertheless, improving model accuracy and reducing prediction errors require the use of hybrid approaches and higher-quality data. The findings of this research can assist urban policymakers in formulating optimal strategies for sustainable development.
Funding
There is no funding support.
Authors’ Contribution
Authors contributed equally to the conceptualization and writing of the article. All of the authors approved thecontent of the manuscript and agreed on all aspects of the work declaration of competing interest none.
Conflict of Interest
Authors declared no conflict of interest.
Acknowledgments
We are grateful to all the scientific consultants of this paper.
A B S T R A C T
Urban development is one of the most dynamic socio-economic and environmental processes, with far-reaching impacts on human societies and the environment. Analyzing the directions of urban expansion—especially in areas experiencing rapid urban growth and demographic changes—is crucial for efficient resource management and sustainable planning. This study aims to examine the role of machine learning algorithms in analyzing and predicting urban development trends, with a particular focus on their integration with Geographic Information Systems (GIS) and satellite imagery. A systematic search was conducted in databases such as Google Scholar and Scopus, covering the period from 2006 to 2024. From the reviewed literature, 83 articles were selected and analyzed. MAXQDA software was used for data analysis, enabling the identification and classification of key concepts. The selection of articles was based on criteria such as journal credibility, research methodology, and relevance to the topic. The data were coded and categorized using qualitative content analysis in MAXQDA. The results indicated that integrating machine learning algorithms with GIS improved the accuracy of urban development prediction by 10% to 40% compared to traditional models. Among the studied methods, the Random Forest model outperformed Support Vector Machines (SVM) and Artificial Neural Networks (ANN). This study, through a systematic review and comparative analysis of machine learning algorithms, demonstrates that combining these methods with spatial data and satellite imagery can significantly enhance the accuracy of urban development modeling.
کلیدواژهها [English]